Greater exposure to estrogen throughout a woman's lifetime increases her risk of developing breast cancer. In the gut, microflora play a significant role in the metabolism of estrogens. Therefore, inter-individual differences in host bacterial populations may be a determinant of estrogen exposure and ultimately of breast cancer risk. Colonic microfloral production of equol from the soy isoflavone daidzein serves as a biomarker of a unique intestinal bacterial population. Evidence from several studies suggests that, irrespective of soy intake, women with the capacity to produce equol have hormonal profiles associated with a lower risk of breast cancer. Only about one third of individuals have the yet-to-be-identified bacteria capable of producing equol, and equol-producer status can be determined readily from a urine sample collected after a 3-day soy challenge. To date, there has been no systematic study of the effect of equol producer phenotype on estrogen-dependent biomarkers. We propose to examine the association between equol producer phenotype and current and lifetime measures of reproductive hormone exposure. Circulating reproductive hormones and urinary estrogen metabolites will be examined by equol producer phenotype to evaluate current hormonal exposure. Breast and bone densities (markers of lifetime estrogen exposure) will be examined by equol producer phenotype. We will recruit healthy female volunteers, aged 50 to 75 years, from among the 173 women who participated in the Physical Activity and Total Health Study (NCI R01 CA69334; Effect of Exercise on Sex Hormones in Postmenopausal Women). Women in this study had a baseline mammogram to determine percent breast density by computer-assisted technology, total bone density measured by DEXA, a blood sample collected for the analysis of circulating reproductive hormones, and a urine sample collected for the analysis of estrogen metabolites. Thus, by using this study population, we make efficient use of a comprehensive compilation of existing breast density, bone density, and circulating and urinary hormone data. Each woman will complete a 3-day soy challenge and collect the first-void urine on the fourth day for analysis of equol. We will classify women as equol producers or non-producers and will examine differences in percent breast density by equol producer phenotype. In addition, we will establish whether equot production predicts differences in the measures of hormonal exposure, independently of other factors known to be associated with these measures. The results of this study will provide novel data regarding the relationship of the equol-producer phenotype, a marker of colonic nticrofloral environment, to lifetime estrogen exposure.